This paper presents a method to deal with the illposed and rank-deficient linear systems arising from accelerated partial element equivalent circuit-based electromagnetic simulations via a reluctance method. Since conventional, direct methods, cannot be used to solve these kind of problems, regularization techniques need to be employed. Among various regularization techniques, a least square-based method entitled LSQR is utilized to solve the rank-deficient problems. The proposed method is specially proper for the models where capacitive couplings can be neglected, since magnetic field is the dominating factor, like problems within power electronics area. The correctness of the presented PEEC-based solver is ensured by studying bus bar models which are a part of a frequency converters with the application in power electronics.

This paper presents an efficient modeling approach for large bus bar systems in power electronic frequency converters using partial element equivalent circuit (PEEC) method and circuit-level simulations. Several acceleration methods have been applied to improve the computational speed and memory efficiency. The approach is verified against measurements, and used to analyze the impact of the stray impedance of bus bars on the electromagnetic and circuit behavior of a static frequency converter system.

No abstract

This paper presents recent progress in t PEEC based electromagnetic simulations the design of complex bus bar structures level power converters. The approach pre different dedicated acceleration methods design tasks. The first acceleration techn so called reluctance matrix method for fu reducing memory consumption by orders computing time by a factor 3 to 5. The s method applied is based on model techniques for port-to-port impedance ex the computation time by about one order allowing wideband macro modeling simulations. The paper focuses on the ap methods showing the impact on practic tasks.

Abstract-A parallel implementation of a quasi-static Partial Element Equivalent Circuit (PEEC)-based solver that can handle electromagnetic problems with non-orthogonal structures is presented in this paper. The solver has been written in C++ and employs GMM++ and ScaLAPACK computational libraries to make the solver fast, efficient, and adaptable to current parallel computer systems. The parallel PEEC-based solver has been tested and studied on high performance computing clusters and the correctness of the solver has been verified by doing comparisons between results from orthogonal routines and also another type of electromagnetic solver, namely FEKO. Two non-orthogonal numerical test cases have been analysed in the time and frequency domain. The results are given for solution time and memory consumption while bottlenecks are pointed out and discussed. The benchmarks show a good speedup which gets improved as the problem size is increased. With the capability of the presented solver, the non-orthogonal PEEC formulation is a viable tool for modelling geometrically complex problems.